Summary
Tranexamic acid is an effective treatment to reduce blood loss. We performed a retrospective observational study to evaluate safety in unilateral total knee arthroplasty. We utilised Taiwan's national health insurance database to identify relevant patients and to retrieve information on peri‐operative blood transfusions and tranexamic acid administration within 60 days of follow‐up. We examined changes in the rate of transfusions and adverse events with respect to tranexamic acid administration using logistic regression. We observed a total of 226,719 knee arthroplasty cases during 2010–2019. Transfusion and tranexamic acid administration rates were 38.9% (88,258) and 42.9% (97,237), respectively. Tranexamic acid was associated with a 50% decrease in blood transfusions (RR: 0.50, 95%CI: 0.48–0.51). After propensity‐score matching, tranexamic acid was not associated with pulmonary embolism; deep vein thromboembolism; artery vein thromboembolism; acute myocardial infarction; ischaemic stroke; or in‐hospital mortality, but was significantly associated with acute kidney injury. Patients with existing chronic kidney disease suffered a high absolute risk of kidney injury irrespective of tranexamic acid administration (832 per 10,000, 95%CI 797–869). Tranexamic acid was also associated with surgical site infection. There was strong interaction between blood transfusion; tranexamic aid administration; and development of surgical site infection. In conclusion, tranexamic acid use was associated with decreased blood transfusion and was not associated with thromboembolic events. However, careful consideration is required before use in patients with pre‐existing renal disease. Further, our observed interaction between patients given tranexamic acid who subsequently require transfusion requires careful consideration with respect to enhanced prophylaxis against surgical site infection.